Atrial fibrillation (AF) has a currently estimated prevalence of 2.3 million people and increases stroke risk by 3-5 fold, heart failure risk by 3 fold and mortality risk by 1.5 to 1.9 times. The study will integrate P wave indices, a novel electrocardiographic (ECG) phenotype, and determine their contribution to risk prediction for AF, stroke and mortality. P wave indices, described by P wave duration, dispersion, amplitude, area, and terminal force, reflect atrial electrophysiology and morphology and are consequently a barometer of atrial tissue function. The study has three aims. First, it will develop the phenotype of P wave indices in the Framingham Heart Study (FHS). Second, it will assess the genetic variants associated with P wave indices in the FHS and the Atherosclerosis Risk in Communities (ARIC) study. Third, it will develop and calibrate risk models for AF, heart failure and mortality across these two cohorts. Models will be developed in ARIC and then validated and replicated in FHS. The study design facilitates collaboration between cohorts with complementary demographics and phenotypic characterization, and leverages previously established collaborations for AF risk prediction. The epidemiology, prognostic value and genetic determinants of P wave indices have been incompletely described. Our efforts will contribute towards understanding risk modeling for AF, heart failure, and mortality within the ARIC and FHS studies. The study will make unique contributions towards the association of P wave indices and the end- points. Ultimately, such an effort will elucidate how P wave indices, easily accessible from the surface ECG, may indicate important physiologic processes that merit intervention. 2/5/2010